Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2016 May 1.
Published in final edited form as: Toxicol Appl Pharmacol. 2015 Mar 7;284(3):330–338. doi: 10.1016/j.taap.2015.02.025

Sex-specific patterns and deregulation of endocrine pathways in the gene expression profiles of Bangladeshi adults exposed to arsenic contaminated drinking water1

Alexandra Muñoz 1, Yana Chervona 1, Megan Hall 2, Thomas Kluz 1, Mary V Gamble 3, Max Costa 1
PMCID: PMC4410068  NIHMSID: NIHMS670646  PMID: 25759245

Abstract

Arsenic contamination of drinking water occurs globally and is associated with numerous diseases including skin, lung and bladder cancers, and cardiovascular disease. Recent research indicates that arsenic may be an endocrine disruptor. This study was conducted to evaluate the nature of gene expression changes among males and females exposed to arsenic contaminated water in Bangladesh at high and low doses. Twenty-nine (55% male) Bangladeshi adults with water arsenic exposure ranging from 50–1000 µg/ L were selected from the Folic Acid Creatinine Trial. RNA was extracted from peripheral blood mononuclear cells for gene expression profiling using Affymetrix 1.0 ST arrays. Differentially expressed genes were assessed between high and low exposure groups for males and females separately and findings were validated using quantitative real-time PCR. There were 534 and 645 differentially expressed genes (p<0.05) in the peripheral blood mononuclear cells of males and females, respectively, when high and low water arsenic exposure groups were compared. Only 43 genes overlapped between the two sexes, with 29 changing in opposite directions. Despite the difference in gene sets both males and females exhibited common biological changes including deregulation of 17β-hydroxysteroid dehydrogenase enzymes, deregulation of genes downstream of Sp1 (specificity protein 1) transcription factor, and prediction of estrogen receptor alpha as a key hub in cardiovascular networks. Arsenic-exposed adults exhibit sex-specific gene expression profiles that implicate involvement of the endocrine system. Due to arsenic’s possible role as an endocrine disruptor, exposure thresholds for arsenic may require different parameters for males and females.

Keywords: arsenic, humans, HSD17B-1, cardiovascular disease

Introduction

Inorganic arsenic (iAs), a known human carcinogen, naturally contaminates drinking water and currently affects approximately 150 million people in at least 70 countries throughout the world (IARC, 2004; Ravenscroft et al., 2009). Exposure to As is associated with skin, lung, bladder, liver and prostate cancers (Benbrahim-Tallaa and Waalkes, 2008; Rahman et al., 2009), cardiovascular disease (Chen et al., 2013; Tseng et al., 2003), diabetes (Pan et al., 2013), and hypertension (Chen et al., 1995). The mechanisms that underlie As’s diverse pathogenesis remain unclear and are thought to include oxidative stress, endocrine disruption, alteration of cell signaling and proliferation processes, and impairment of DNA damage responses (Andrew et al., 2003; Davey et al., 2007; Flora, 2011; Schuhmacher-Wolz et al., 2009). In addition, there is a growing body of evidence to indicate that As acts through epigenetic mechanisms as part of its pathogenesis by altering the abundance of histone modifications and the level of DNA methylation (Reichard and Puga, 2010; Ren et al., 2011).

The evidence to suggest that As acts as an endocrine disruptor includes in vitro activation of steroid hormone receptors (Bodwell et al., 2006; Davey et al., 2007), animal models with altered fertility (Chang et al., 2007; Jana et al., 2006) and sex-specific disease distributions (Waalkes et al., 2003), as well as human studies indicating altered fertility (Hsieh et al., 2008; Xu et al., 2012), sex-specific disease distributions (Aballay et al., 2012), and sex-linked changes in methylation profiles (Pilsner et al., 2012). Previous research from our lab supports this trend with the observed sex-specific changes in the abundance of histone modifications in Bangladeshi adults exposed to As contaminated drinking water (Chervona et al., 2012). Microarray studies that have assessed gene expression changes in response to As contaminated water have evaluated changes with males and females pooled together (Andrew et al., 2008; Lu et al., 2001; Wu et al., 2003) or with only females (Wu et al., 2003). However, given the potential endocrine activity of As, in particular its possible role as a xenoestrogen (Jana et al., 2006; Waalkes et al., 2004), division of sexes in assessing its impact may be necessary to understand its pathogenesis. Indeed a recent study conducted among Flemish adults exposed to an array of toxins, including xenoestrogens, exhibited sex-specific transcriptomic responses (De Coster et al., 2013) that were revealed when sexes were evaluated separately.

Here we evaluate the gene expression profiles of RNA from the peripheral blood mononuclear cells (PBMC) of Bangladeshi adults (n=29) chronically exposed to As contaminated drinking water. Subjects with high water As exposure are compared to those with low water As exposure among males and females separately. To evaluate the sex-specific profiles we compare the areas of biological relevance for each sex and evaluate to what extent the same genes are found among biologically relevant areas that are common to males and females.

Materials and Methods

Study Site and Subject Recruitment

The Health Effects of Arsenic Longitudinal Study (HEALS) cohort is part of Columbia University’s Superfund Research Program and currently includes approximately 30,000 participants (Ahsan et al., 2006). The present study utilizes samples from a subset of 600 HEALS participants who were recruited for enrollment into the Folic Acid and Creatine Trial (FACT), a clinical trial that aimed to lower blood As concentrations with nutritional supplementation and therefore required that participants be drinking from wells with water As concentrations greater than or equal to 50 µg/L. Fieldwork was completed in June 2011. Study participants were given a water filtration system for removal of As at the time of enrollment. Participants with renal or gastrointestinal diseases, those taking nutritional supplements, or who were pregnant and/or planning to become pregnant were not included in the study. This study evaluates the gene expression patterns of 29 study participant from the FACT clinical trial. Our study population is comprised of adults ranging in age from 26–53. Here the effects of relatively low exposure (50–200 µg/L) to water arsenic are compared to high exposure (250–1000 µg/L) for males (n=16) and females (n=13) separately. Among the males there were 9 males with low exposure and 7 with high exposure. Among the females there were 6 females with low exposure and 7 with high exposure. Please see table 1 of the results for other information regarding the demographic features of the groups.

Table 1.

General characteristics for the total study sample by sex and water As groupa

Total sample
(n=29)		 Males,
Low water
Asb
(n=9)		 Males,
High water
Asc
(n=7)		 Females,
Low water
Asd
(n=6)		 Females,
High
water Ase
(n=7)
Age 40.0 ± 7.2 42.6 ± 6.4 41.7 ± 6.8 36.7 ± 9.6 37.7 ± 5.6
BMI, kg/m2 20.4 ± 2.9 19.7 ± 2.1 18.2 ± 1.1 23.8 ± 2.1 20.6 ± 3.0
Education, years 2.9 ± 3.7 2.8 ± 3.4 0 5.0 ± 4.5 4.0 ± 3.8
water As µg/L 246.4 ± 207.4 103.0 ± 33.7 354.5 ± 136.2 116.7 ± 61.9 433.9 ± 274.2
Urinary As, µg/L 264.1 ± 349.3 116.8 ± 76.4 456.3 ± 597.3 113.3 ± 99.7 390.6 ± 281.7
Urinary creatinine (mg/dL) 75.0 ± 41.5 75.3 ± 40.9 58.0 ± 38.8 74.5 ± 54.7 91.9 ± 33.9
Urinary As/gm creatinine 345.8 ± 324.9 165.5 ± 74.9 650.2 ± 450.7 178.9 ± 186.7 416.3 ± 250.1
Blood As (µg/L) 11.4 ± 10.3 6.8 ± 3.1 20.7 ± 15.3 7.1 ± 7.2 11.7 ± 7.3
Male, % 55.2 - - - -
Land Ownership, % 37.9 44.4 42.9 33.3 28.6
Television Ownership, % 24.1 11.1 14.3 33.3 42.9
Current cigarette smoking, % 44.8 77.8 85.7 0 0
Current betel nut Use, % 34.5 55.6 42.9 16.7 14.3
Open in a new tab
a

mean ± SD unless otherwise noted,

b

water As 50 to 150 µg/L,

c

water As 232 to 500 µg/L,

d

water As 50 to 200 µg/L,

e

water As 250 to 1000 µg/L

Sample collection and handling

Blood samples were obtained by venipuncture, and collected into EDTA-containing vacutainer tubes, which were then placed in IsoRack/IsoPack cool packs (Brinkmann Instruments). Samples were transported in hand-carried coolers to the local laboratory at the field clinic in Araihazar, within four hours of collection. Samples were centrifuged for 10 minutes at 3,000 × g at 4°C, and the plasma fraction was stored at −80°C. PBS was added to the remaining cells followed by a ficoll-hypaque gradient extraction of PBMCs using standard techniques. PBMCs were stored at −80°C. Urine samples were collected with 50-mL acid-washed polypropylene tubes. Within 4 hours samples were transported from portable coolers to freezers (−20°C) in the Araihazar laboratory, and were then hand carried on dry ice to New York.

Water and urinary As

Water sample collection, sample handling, and analysis for this study were performed as previously described (Chervona et al., 2012). The water As samples were analyzed at Columbia University’s Lamont Doherty Earth Observatory by inductively coupled mass spectrometry (ICP-MS), with a detection limit of 0.1 µg/L. Total urinary As (uAs) concentrations were measured at Columbia University Trace Metals Core Lab by GFAA spectrometry using an AAnalyst 600 graphite furnace system (PerkinElmer), as described (Nixon et al., 1991). This laboratory is part of a quality control program run by the Institut de Sante Publique du Quebec, Canada. Intraclass correlation coefficients (ICCs) were 0.99 between the Columbia laboratory’s values and samples calibrated at the Quebec laboratory. To correct for hydration status, urinary creatinine was analyzed using a method based on the Jaffe reaction.

RNA isolation, amplification, and hybridization

Total RNA was extracted from each sample according to the TRIzol (Invitrogen) manufacturer's protocol, and purified using RNeasy Plus Micro Kit (Qiagen). Sense strand cDNA probes were synthesized (amplified) using Ambion Whole Transcript Expression Kit. Amplified single stranded cDNA was fragmented and labeled using GeneChip WT Terminal Labeling Kit (Affymetrix). The fragmented and labeled DNA underwent hybridization with Affymetrix GeneChip Human Gene 1.0 ST Array that contains 28,869 well-annotated genes. Samples were processed in two batches with representatives from each exposure category in each batch and were randomized during processing.

Data analysis to identify differentially expressed genes

Gene expression analyses were performed using R. Gene expression data were imported and normalized using a single-sample method, single channel array normalization (SCAN) (Piccolo et al., 2012), which summarizes probes to Entrez IDs, in R 2.15.0 GUI 1.53 64-bit. Data were processed to expression sets using the Affymetrix package version 1.30.0. in R 2.15.1 GUI 1.42 Leopard build 64-bit, and were filtered to remove negative expression values as per SCAN protocol. Unsupervised hierarchical clustering using the Pearson correlation method was used to evaluate potential batch effects. Surrogate variable analysis was used to estimate seven surrogate variables and performed in order to remove batch effects and other sources of expression heterogeneity (Leek and Storey, 2007). Seven surrogate variables were incorporated into the analysis and such incorporation was used in lieu of correcting for demographic features of the population, which was not possible due to the small sample size. Significance of gene expression changes for males and females were assessed separately between high and low water As exposure groups using a gene-wise linear model approach with LIMMA 3.14.4 (Smyth, 2004). LIMMA utilizes an empirical Bayes approach to generate moderated t-statistics by taking into account the standard errors and estimated log-fold changes. Probe summarization and filtration occurred prior to LIMMA analysis and therefore 11623 genes were tested for significance using moderated t-statistics and a p-value threshold of p < 0.05.

Heat maps were generated using the heatmap.2 function in R and unsupervised hierarchical clustering was governed by Pearson correlation method. Networks and functional analysis were generated using IPA (Ingenuity ® Systems, www.ingenuity.com). Gene set enrichment analysis was performed using the molecular signatures database (Broad Institute Gene Set Enrichment Analysis, http://www.broadinstitute.org/gsea). To evaluate the influence of cigarette smoking, lists of differentially expressed genes were evaluated for the presence of known biomarkers of cigarette smoking (van Leeuwen et al., 2005) and statistical significance of biomarker representation was assessed using a hypergeometric distribution as previously described (Draghici et al., 2003). All microarray data is MIAME compliant and the raw data has been deposited in NCBIs Gene Expression Omnibus (GEO), and assigned series accession number GSE57711.

Quantitative real-time PCR

Gene expression changes were validated using quantitative real-time polymerase chain reaction (q-RT-PCR). Total RNA was extracted from each sample using TRIzol Reagent (Invitrogen), and converted to single stranded cDNA using Superscript III (Invitrogen). q-RT-PCR analysis was performed using SYBR green PCR system (Applied Biosystems) on ABI prism 7900HT system (Applied Biosystems). All q-RT-PCR reactions were performed in triplicate. Relative gene expression level, normalized to 18s rRNA expression, was presented as median relative mRNA values ± SD (n=9) for participants in low and high exposure groups. Statistical significance was tested using two-tailed, unpaired t-test. The expression of the target genes was first normalized to the expression of the house-keeping gene glyceraldehyde 3-phosphate dehydrogenase (GADPH).

Results

Demographic data

Study population characteristics are presented in Table 1 for the study population by water As exposure category (low vs. high). Briefly, a total of 29 participants were selected for this study with 16 males and 13 females. Among the males the average age was 42.6 ± 6.4 for the males with low exposure and 41.7± 6.8 for the males with low exposure. Among the females the average age was 36.7± 9.6 for females with low exposure and 37.7 ± 5.6 for the females with high exposure. The median water As concentration for the low exposure group was 103 µg/L for males and 117 µg/L for females (range 50–200 µg/L). For the high exposure group, the median water As concentration was 355 µg /L for males (range 250–500 µg /L) and 434 µg/L for females (range 232–1000 µg /L).

Gene expression analysis

In the PBMC’s of highly exposed males, there were 534 differentially expressed genes (p < 0.05), when compared to males with low As exposure, with 271 up-regulated and 264 down-regulated. For the highly exposed females there were 645 differentially expressed genes (p < 0.05) when compared to females with low water As exposure, with 303 up-regulated and 342 down-regulated. The unsupervised hierarchical clustering analysis for differentially expressed genes of males (Figure 1a) and females indicates clear clustering based on water As exposure group (low vs. high) for males and females (Figure 1b), with the exception of one outlier for females, indicating that dose is driving clustering in our lists. Only 43 genes overlapped between the two sexes, with 29 out of 43 genes changing in opposite directions (Figure 1C; Table 2). Biological relevance of gene expression was assessed via network analysis using IPA and top ranked networks and enrichment scores are presented in Table 3.

Figure 1. Gene Expression Profiles of As Exposed Adults.

Figure 1

Open in a new tab

Unsupervised hierarchical cluster analysis and heatmap of differentially expressed genes (p < 0.05) for males (A) and females (B) with chronic exposure to As contaminated drinking water at high (H) and low (L) doses. C) Unsupervised hierarchical cluster analysis and heatmap of differentially expressed genes represented among males and females (p < 0.05). Each row represents a statistically significant gene determined via LIMMA. Samples (columns) are labeled according to the following codes with appended numbers corresponding to unique individuals: ML - males with low water As exposure, MH - males with high water As exposure, FL- females with low water As exposure, FH - females with high water As exposure.

Table 2.

Differentially expressed genes common to both malesa and femalesb

Gene
Symbol		 Description Absolute fold change
Malesa Femalesb
SLC37A2 solute carrier family 37 (glycerol-3- phosphate transporter), member 2 1.12987 −1.05461
ZNF429 zinc finger protein 429 1.11314 −1.09785
ABCA1 ATP-binding cassette, sub-family A (ABC1), member 1 1.09686 −1.11128
KMO kynurenine 3- monooxygenase (kynurenine 3-hydroxylase) 1.09616 −1.11952
CISH cytokine inducible SH2-containing protein 1.08392 1.10739
TUBA1B tubulin, alpha 1b 1.07894 1.06552
PKIG protein kinase (cAMP-dependent, catalytic) inhibitor gamma 1.07413 −1.06864
SEMA4A sema domain, immunoglobulin domain (Ig), transmembrane domain (TM) and short cytoplasmic domain, (semaphorin) 4A 1.06561 1.08562
CLCN4 chloride channel, voltage-sensitive 4 1.06541 −1.08737
RBBP8 retinoblastoma binding protein 8 1.06036 −1.07956
TMEM68 transmembrane protein 68 1.05458 −1.08187
IFIH1 interferon induced with helicase C domain 1 1.05346 −1.10904
GLB1 galactosidase, beta 1 1.04594 1.04002
BTF3P11 basic transcription factor 3 pseudogene 11 1.04533 1.04236
VKORC1L1 vitamin K epoxide reductase complex, subunit 1-like 1 1.04435 −1.04563
NOL11 nucleolar protein 11 1.04418 −1.05934
MRPS18B mitochondrial ribosomal protein S18B 1.04346 −1.03351
SNORA29 small nucleolar RNA, H/ACA box 29 1.04276 1.04468
PRMT10 protein arginine methyltransferase 10 (putative) 1.04159 −1.04801
HSF5 heat shock transcription factor family member 5 1.03604 1.04256
PQLC1 PQ loop repeat containing 1 1.03523 1.04041
KRTAP5-1 keratin associated protein 5-1 1.03504 1.03366
MRPS22 mitochondrial ribosomal protein S22 1.03337 −1.04741
MAD2L1BP MAD2L1 binding protein 1.03253 −1.04613
SMAP1 small ArfGAP 1 1.02688 1.03809
MAP3K6 mitogen-activated protein kinase kinase kinase 6 −1.02623 1.03868
ZFPL1 zinc finger protein-like 1 −1.02681 −1.036
KBTBD3 kelch repeat and BTB (POZ) domain containing 3 −1.03002 −1.03796
RNF126 ring finger protein 126 −1.03488 1.04559
PPP1R12B protein phosphatase 1, regulatory subunit 12B −1.03503 1.0613
RORA RAR-related orphan receptor A −1.0373 1.05826
SLC11A2 solute carrier family 11 (proton-coupled divalent metal ion transporters), member 2 −1.04141 1.05051
SNTB1 syntrophin, beta 1 (dystrophin-associated protein A1, 59kDa, basic component 1) −1.04204 1.06052
HERC2 HECT and RLD domain containing E3 ubiquitin protein ligase 2 −1.04328 1.04903
TNIK TRAF2 and NCK interacting kinase −1.04349 1.04896
ATIC 5-aminoimidazole-4- carboxamide ribonucleotide formyltransferase/IMP cyclohydrolase −1.04567 1.04872
IL11RA interleukin 11 receptor, alpha −1.04633 1.1028
ETS2 v-ets erythroblastosis virus E26 oncogene homolog 2 (avian) −1.04934 1.06649
ACER2 alkaline ceramidase 2 −1.05118 −1.06698
CLYBL citrate lyase beta like −1.06358 1.07364
CCDC144NL coiled-coil domain containing 144 family, N-terminal like −1.065 −1.05019
PRDM1 PR domain containing 1, with ZNF domain −1.06919 1.08734
KIAA1324 KIAA1324 −1.09944 1.14391
Open in a new tab
a

534 differentially expressed genes (p<0.05) for males high water As vs. males low water As

b

645 differentially expressed genes (p<0.05) for females high water As vs. females low water As

c

expression values for males high water As relative to males low water As

d

expression values for females high water As relative to females low water As

Table 3.

Top IPA networks for differentially expressed genes of males (males high water As vs. males low water As) and females (females high water As vs. females low water As)

Network Score Network Score
Males Females
Post-Translational Modification, Protein Folding, Antimicrobial Response 75 Cardiovascular Disease, Cellular Compromise, Neurological Disease 67
Cardiovascular Disease, Congenital Heart Anomaly, Developmental Disorder 66 Cell Death and Survival, Hematological System Development and Function, Tissue Morphology 42
Amino Acid Metabolism, Molecular Transport, Small Molecule Biochemistry 62 Molecular Transport, RNA Trafficking, Cellular Development 35
Lipid Metabolism, Small Molecule Biochemistry, Cell Morphology 48 DNA Replication, Recombination, Repair, Cell Morphology, Cellular Function and Maintenance 25
Cellular Development, Cellular Growth and Proliferation, Hematological System Development 48 Cellular Function and Maintenance, Cell Death and Survival, Tissue Morphology 24
Open in a new tab

Differentially expressed genes were also evaluated for highly ranked gene signatures using GSEA database and the top ranked signatures from all possible signatures to query are presented in Supplementary Material, Table S1. Cardiovascular disease was a primary factor in a top-ranked network for both males and females, and each network was comprised of unique gene lists that are presented in Supplemental Material, Tables S2 and S3, with only one gene in common (RBBP8 (retinoblastoma-binding protein 8)) between the sexes. Genes responsive to Sp1 (specificity protein 1) transcription factor comprised a top-ranked gene signature for both males and females. Sp1 responsive genes were evaluated for their similarity between the sexes. Males had 110 Sp1 regulated genes and females had 91, both with a 50/50 split between up and down-regulated genes. There was little overlap between the lists with only 9 genes overlapping, which are presented in Supplementary Material, Table S4. Due to the fact that the majority of males were smokers and no females were reported as smokers (Table 1), we screened the male and female gene sets for 26 genes that exhibited 90% sensitivity in predicting smokers versus nonsmokers in order to evaluate whether smoking status was confounding the sex-effect (Lampe et al., 2004). For the males, three of the smoking biomarker genes appeared on our gene lists, which was statistically significant (p < 0.0295); and there were no smoking genes in the female gene lists. The smoking genes found in the male gene set included (CYP1B1 (cytochrome P450, family 1, subfamily B, polypeptide 1), EPB41L3 (erythrocyte membrane protein band 4.1-like 3), and BRD3 (bromodomain containing 3)). To further assess the relationship, we probed differentially expressed genes of high males vs. high females, and 3 smoking genes were present in this gene set. While a comparison of low males vs. low females, contained no smoking biomarker genes. Collectively these findings indicate that smoking status is primarily an influence in the high males, but its signature is not dominant as only three smoking biomarker genes are present (CYP1B1, EPB4IL3, IL1B) and clustering in heatmaps is driven by As dose not smoking status.

Real-time PCR Validation

To validate the results obtained from the microarray study, quantitative real-time PCR was performed on a subset of genes with biological relevance. POLR2E (polymerase (RNA) II (DNA directed) polypeptide E, 25kDa), CKS2 (CDC28 protein kinase regulatory subunit 2), and TGM2 (transglutaminase 2 (C polypeptide, protein-glutamine-gamma-glutamyltransferase)) were validated from the male differentially expressed genes. E2F5 (E2F transcription factor 5, p130-binding) was validated from the female differentially expressed genes. As shown in Table 4 the up- or down-regulated patterns for all genes obtained from real-time PCR were consistent with those obtained in the microarray study.

Table 4.

Validation of microarray results by quantitative real-time PCR

Gene Affymetrix fold change Q-RT-PCR fold change
POLRE2 1.12 4.31
CKS2 −1.06 −1.96
TGM2 1.06 1.87
E2F5 −1.10 −1.44
Open in a new tab

POLR2E (polymerase (RNA) II (DNA directed) polypeptide E, 25kDa), CKS2 (CDC28 protein kinase regulatory subunit 2), and TGM2 (transglutaminase 2 (C polypeptide, protein-glutamine-gamma-glutamyltransferase)) were validated in the males. E2F5 (E2F transcription factor 5, p130-binding) was validated in the females. Patterns of fold change found in quantitative realtime PCR exhibit consistent directionality to those obtained by microarray.

Discussion

Our findings suggest that As exposure deregulates expression of genes involved in the immune system, heat shock response and DNA repair processes, which is consistent with previous human studies on this topic. The novelty of our paper is that we find evidence to support the notion that As acts through endocrine associated pathways in both sexes, and that such pathways involve mediation of responses through sex-specific mechanisms as evidenced by the distinct gene sets for males and females.

Given that the exposure inclusion criteria of the FACT study design required that As in drinking water exceed 50 µg/L, we have conducted a comparison of high and low exposure groups. This study design is consistent with previous studies of this cohort that detected changes in the abundance of histone marks in response to high and low levels of As contaminated drinking water (Chervona et al., 2012). Due to the comparison between high and low groups, the range of As exposure represented within each group, and the compression of expression values that occurs when the SCAN normalization protocol is employed, our fold change values are tightly expressed with maximum values in the range of ±1.3. In turn we focus on evaluating the biological significance of the differentially expressed genes, highlighting genes and areas of biological activity that are “deregulated” across exposure categories, and evaluating how such deregulation manifests differently in males and females.

As’s ability to impact gene expression changes is consistent with previous studies which demonstrated that As deregulates the expression of heat shock, DNA repair, and immunoregulatory genes. We briefly mention such changes here to biologically validate our gene lists and contextualize our findings with other similar studies. Heat shock proteins are consistently found to be aberrantly expressed in response to As. Here, HSP1AB, HSPBP1, and HSFS are deregulated in males, and HSPA13 is deregulated in females (Table 5). Our findings are consistent with Andrew et al. and Argos et al. who find HSP1AB is deregulated in their male dominant and female never smoker cohorts, respectively (Andrew et al., 2008; Argos et al., 2006). As’s ability to interfere with DNA repair and promote carcinogenesis as a co-mutagen has been firmly established in experimental systems (Hartwig et al., 1997; Rossman et al., 2004). We find that drinking water As exposure deregulates expression of DNA repair genes in both males and females and includes genes that are involved in excision repair (LIG3, SMUG1), DNA repair regulation (RBBP8, MDC1), and mismatch repair (MSH2) (Table 5). Previous drinking water exposure studies have found that As interferes with nucleotide excision repair (Andrew et al., 2006; Andrew et al., 2003) as well with base excision repair and strand breaks (Wu et al., 2003). With respect to the immune system we find As deregulates a number of interleukins (IL) in both sexes (Table 5), that there was a viral-response like signature in the males (RSAD2 IFIT1 IFIT3, IFIT5), and deregulation of major histocompatibility (MHC) class genes (HLA-DRB6, HLA-B, HLA-G) in the females. Our findings are consistent with previous studies, as Andrew et al. also found deregulation of MHC II class genes in their male dominant cohort (Andrew et al., 2006) and Wu et al. and Argos et al. report deregulation of interleukins. Of those IL deregulated, IL16 (females) was previously reported by Wu et al. in females exposed to As contaminated water in Taiwan where it exhibited dose-responsiveness (Wu et al., 2003) and IL1RN (males) was reported by Argos et al. to be down-regulated in women with Asal induced skin lesions (Argos et al., 2006).

Table 5.

Biological validation of gene lists. Differentially expressed genes involved in heatshock response, DNA repair, and immune response for males and females.

Gene symbol Description Fold change
Heat shock
    Males
HSP1AB heat shock 70kDa protein 1B 1.055
HSPBP1 heat shock 70kDa biding protein, cytoplasmic chaperone 1 1.038
HSF5 heat shock transcription family member 5 1.036
    Females
HSPA13 heat shock 70kDa protein 13 −1.054
DNA repair
    Males
RBBP8 retinoblastoma binding protein 8 1.060
CDK7 cyclin-dependent kinase 7 1.058
POLD3 polymerase (DNA-directed), delta 3, accessory subunit 1.036
POLI polymerase (DNA directed) iota), RPA3 (replication protein A3, 14kDa −1.054
RPA3 replication protein A3, 14kDa 1.030
LIG3 ligase III, DNA, ATP-dependent −1.027
    Females
RBBP8 retinoblastoma binding protein 8 −1.080
MDC1 mediator of DNA-damage checkpoint 1 1.041
LIG4 ligase IV, DNA, ATP-dependent −1.060
MSH2 mutS homolog 2, colon cancer, nonpolyposis type 1 (E. coli) −1.059
SMUG1 single-strand-selective monofunctional uracil-DNA glycosylase 1 −1.069
Immune system
    Males
RSAD2 S-adenosyl methionine domain containing protein 2 1.186
IFIT1 interferon-induced protein with tetratricopeptide repeats 1 1.207
IFIT3 interferon-induced protein with tetratricopeptide repeats 3 1.123
IFIT5 interferon-induced protein with tetratricopeptide repeats 5 1.095
IFI44L interferon-induced protein 44-like 1.323
IFI6 interferon, alpha-inducible protein 6 1.201
MX1 myxovirus (influenza virus) resistance 1, interferon-inducible protein p78 (mouse) 1.180
IL23A interleukin 23, alpha subunit p19 −1.045
IL11RA interleukin 11 receptor, alpha −1.046
IL15RA interleukin 5 receptor, alpha −1.047
IL1RL1 interleukin 1 receptor-like 1 −1.102
IL1RN interleukin 1 receptor antagonist −1.084
IL17C interleukin 17C −1.037
    Females
IL11RA interleukin 11 receptor, alpha 1.103
IL13RA1 interleukin 13 receptor, alpha 1 1.089
IL18BP IL18 binding protein 1.070
IL16 interleukin 16 −1.035
HLA-DRB6 major histocompatibility complex, class II, DR beta 6 (psuedogene) −1.083
HLA-B major histocompatibility complex, class I, B −1.083
HLA-G major histocompatibility complex, class I, G −1.129
IFIH1 interferon induced with helicase C domain 1 −1.109
Open in a new tab

In addition to the areas discussed above, we observe that As deregulates expression of 17β-hydroxysteroid dehydrogenases (17β-HSDs), which are a family of enzymes involved in regulating steroid hormones throughout the body. Both males and females exhibited deregulation of a HSD family member. In females 17β-hydroxysteroid dehydrogenase type 4 (HSD17B4) was deregulated. This enzyme can oxidize estradiol to estrone and has widespread constitutive expression with the highest levels in the liver where it is involved in fatty acid metabolism and synthesis of bile acids (Moeller and Adamski, 2006). In males HSD17B10 was deregulated, and it exhibits broad substrate specificity for biomolecules including fatty acids, estrogens, androgens, corticosteroids and progestins and is constitutively expressed in the liver, gonads, and brain, where its regional expression levels may play a critical role in neurosteroidogenesis (Moeller and Adamski, 2006; Yang et al., 2005). HSD17B10 has been implicated in the development of Alzheimer’s disease (Moeller and Adamski, 2006), which is a highly ranked gene signature for males and females (Supplementary Material, Table S1). HSD17B10 is also utilized in the alternative pathway for biosynthesis of 5α-dihydrotestosterone (5α-DHT) from adiol (Yang et al., 2005). Deregulation of HSD17B10 may help to explain why males exposed to As exhibit reduced semen quality (Xu et al., 2012) and increased risk of erectile dysfunction (Hsieh et al., 2008). Animal models support this notion and consistently yield decreased levels of testicular enzymes including 17β-HSD in conjunction with declined fertility parameters in both rats and mice exposed to As contaminated drinking water (Chang et al., 2007; Jana et al., 2006; Pant et al., 2004).

For both males and females cardiovascular disease was a top-ranked network, and this finding is consistent with strong epidemiological data implicating As in the progression and development of cardiovascular disease (Argos et al., 2010; Chen et al., 2009). The specific nature of the male and female cardiovascular disease networks was distinct as represented by their unique gene lists (Supplementary Material, Tables S2 and S3), which had only one gene in common (RBBP8), and different secondary network features. Such differences may arise via sex-specific actions of sterol pathways as both networks share the involvement of the estrogen receptor (ER) and luteinizing hormone (LH) as key hubs (Supplementary Material, Figures S2 and S3). Our findings suggest that while cardiovascular disease is a common outcome of long-term As exposure, males and females may have different mechanisms that dominate in the pathogenesis leading to cardiovascular disease. Sex-specific pathogenesis of As-associated cardiovascular disease is also evident in recent work from the HEALS cohort study, which finds a positive association in women only between long-term As exposure and QT prolongation time, a measurement of the electrocardiogram that indicates increased risk for severe cardiac events (Chen et al., 2013).

The top ranked gene signature for both males and females was genes containing transcriptional binding sights for specificity protein 1 (Sp1) transcription factors. As mentioned in the results there were about 100 genes regulated by Sp1 in gene sets of males and females with only 9 genes overlapping between the genders (Supplemental Material, Table S4). Sp1 can positively or negatively regulate gene expression (Li et al., 2004), is deregulated in numerous cancers and neurodegenerative disorders, and is involved in cell cycle regulation, angiogenesis, and senescence (Chang and Hung, 2012), which are all events consistent with As’s pathogenesis. It is unlikely that Sp1 deregulation occurs through direct interaction of As and Sp1 zinc fingers, because recent research indicates that trivalent As only binds zinc finger motifs with more than two cysteine residues, and does not bind to Sp1 which has two (Zhou et al., 2014). However, Sp1 activity is modulated by post-translational modifications including acetylation, methylation, sumoylation, and ubiquitination that regulate Sp1 protein level, transactivation activity, and DNA binding affinity (Chang and Hung, 2012). As may influence Sp1 modifications as it has been shown to alter acetylation of histones (Chervona et al., 2012), methylation of DNA (Ren et al., 2011), and ubiquitination of proteins (Bredfeldt et al., 2004; Kirkpatrick et al., 2003). Such modifications can alter Sp1’s ability to partner with steroid receptors (SR) including ERα, the mineralocorticoid receptor (MR), the glucocorticoid receptor (GR), and the androgen receptor (AR), and thereby alter the synergistic couplings involved in transcription of SP1 or SR responsive genes (Meinel et al., 2013; Ou et al., 2006; Porter et al., 1997; Yuan et al., 2005). In vitro models indicate that As influences transcription of genes downstream of ER, MR, and GR, but the mechanism driving these changes is unknown (Bodwell et al., 2006), and we suggest that post-translational modification of the Sp1 pool may be involved.

The gene expression profiles observed in males and females indicate that similar channels mediate As’s impact in both sexes. However, the striking difference in gene sets for the males and females, suggests that As may exert its influence through sex-specific endocrine interactions, which could lead to sex-specific treatment or prevention strategies. Our data support the growing trend that As can be considered an endocrine disruptor and further study is needed to elucidate how this interaction may occur (Bodwell et al., 2006; Davey et al., 2007), In addition, our data support the perspective that toxic actions may be sex-specific and that evaluation of biomarkers and determination of safety thresholds may require separate analyses for males and females responses (De Coster et al., 2013).

Given the nature of working with a human population, there are a number of study limitations arising from demographic features of the study population that must be addressed before concluding. The smoking status of the male cohort is one demographic feature that may be influencing the presentation of our data. Due to the prevalence of smoking among males in this region of Bangladesh, the majority of males were self-reported smokers while none of the females were reported as smokers. In order to evaluate the effect of smoking status on our results, we probed for detection of genes that are biomarkers for smoking among multiple datasets and found limited detection of such genes. To further evaluate the influence of smoking, we assessed the biological validity of our gene lists to insure that our lists are representative of a response to As exposure rather smoking status. As discussed we found that deregulation of heat shock, DNA repair, and immunoregulatory responses were consistent with changes observed in these areas in similar studies (Andrew et al., 2008; Andrew et al., 2006; Andrew et al., 2003; Argos et al., 2006; Wu et al., 2003), indicating that the lists generated were a reflection of As exposure. The relevance of our gene lists is further supported by the fact that the unsupervised clustering in the heatmap (Figure 1a) is driven by As dose and not smoking status. Collectively, this analysis suggests that the sex-specific profiles exhibited in our data are primarily a reflection of As exposure, which may to a limited extent be influenced by smoking status of males. Another constraint of this study was the limited sample size, which in combination with the fold-change compression previously discussed, effectively inhibited our ability to detect differentially expressed genes using multiple hypothesis correction (i.e. FDR). Other epidemiologic studies conducting gene expression analysis in human populations have also generated uncorrected gene lists, which were utilized for pathway analysis in a similar manner (De Coster et al., 2013; Hebels et al., 2011; Wang et al., 2005). The biological validation performed on our datasets again suggests that the gene lists generated here reflect As exposure rather than randomly selected genes, given their consistency with previous gene expression studies as well as their prediction of diseases well-known to be associated with As exposure such as cardiovascular disease.

Conclusion

Our data indicate that As is influencing gene expression profiles in a manner consistent with previous human gene expression studies in which heat shock proteins, DNA repair processes, and immune responses were deregulated. In addition, we observe sex-specific profiles that may be driven by an interaction between As and the endocrine system. The observed deregulation of HSD enzymes, the presence of the ERα and LH as predicted hubs in the cardiovascular disease networks, and the deregulated Sp1 genes all implicate possible mechanisms through which As may be exerting its multifactorial influence. The involvement of steroid regulating enzymes, the prediction of steroid responsive hubs, and the cooperation between Sp1 and steroid receptors all indicate that As may be acting through endocrine mediated channels and that such effects require separate evaluation in males and females. These findings suggest that exposure thresholds may require different parameters for males and females and that further study is needed to fully understand the impact and nature of sex differences in As pathogenesis.

Supplementary Material

Highlights.

  • Males and females exhibit unique gene expression changes in response to arsenic

  • Only 23 genes are common among the differentially expressed genes for the sexes

  • Male and female gene lists exhibit common biological implications

  • Both sexes exhibit deregulation of cardiovascular and endocrine pathways

Acknowledgments

This work was supported by the following grant numbers: Y. Chervona and M. Gamble from P42 ES10349. M. Costa by R01 ES17875, ES014454, ES005512 ES010344, CA16087 and R01 CA133595 from NCI; and RR029893 from National Center for Research Resources M. Gamble from R01 ES017875 NIEHS, R01 CA133595 from National Cancer Institute (NCI); M. Hall by R00ES018890. T. Kluz by ES000260 NIEHS. A. Munoz by National Science Foundation Graduate Research Fellowship under Grant number 1137475 and NIEHS 5 T32 ES007324. We thank Yu Chen for the helpful comments and editing of the manuscript.

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Disclosure of Potential Conflicts of Interest: The authors have no actual or potential conflicts of interest.

1

Abbreviations: androgen receptor (AR), Arsenic (As), Cardiovascular disease (CVD), estrogen receptor (ER), Folic Acid Creatinine Trial (FACT), glucocorticoid receptor (GR), Health Effects of Arsenic Longitudinal Study (HEALS), 17Β-Hydroxysteroid dehydrogenase (HSD17B), hydroxysteroid dehydrogenase (HSD), single channel array normalization (SCAN), specificity protein 1 (Sp1), steroid receptors (SR), urinary As (uAs), water As (wAs)

References

  1. Aballay L, Díaz M, Francisca F, Muñoz S. Cancer incidence and pattern of arsenic concentration in drinking water wells in Córdoba, Argentina. International journal of environmental health research. 2012;22:220–231. doi: 10.1080/09603123.2011.628792. [DOI] [PubMed] [Google Scholar]
  2. Ahsan H, Chen Y, Parvez F, Argos M, Hussain AI, Momotaj H, Levy D, van Geen A, Howe G, Graziano J. Health Effects of Arsenic Longitudinal Study (HEALS): description of a multidisciplinary epidemiologic investigation. Journal of exposure science & environmental epidemiology. 2006;16:191–205. doi: 10.1038/sj.jea.7500449. [DOI] [PubMed] [Google Scholar]
  3. Andrew A, Jewell D, Mason R, Whitfield M, Moore J, Karagas M. Drinking-water arsenic exposure modulates gene expression in human lymphocytes from a U.S. population. Environmental health perspectives. 2008;116:524–531. doi: 10.1289/ehp.10861. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Andrew AS, Burgess JL, Meza MM, Demidenko E, Waugh MG, Hamilton JW, Karagas MR. Arsenic exposure is associated with decreased DNA repair in vitro and in individuals exposed to drinking water arsenic. Environmental Health Perspectives. 2006;114:1193–1198. doi: 10.1289/ehp.9008. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Andrew AS, Karagas MR, Hamilton JW. Decreased DNA repair gene expression among individuals exposed to arsenic in United States drinking water. Int J Cancer. 2003;104:263–268. doi: 10.1002/ijc.10968. [DOI] [PubMed] [Google Scholar]
  6. Argos M, Kalra T, Rathouz PJ, Chen Y, Pierce B, Parvez F, Islam T, Ahmed A, Rakibuz-Zaman M, Hasan R, et al. Arsenic exposure from drinking water, and allcause and chronic-disease mortalities in Bangladesh (HEALS): a prospective cohort study. Lancet. 2010;376:252–258. doi: 10.1016/S0140-6736(10)60481-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Argos M, Kibriya MG, Parvez F, Jasmine F, Rakibuz-Zaman M, Ahsan H. Gene expression profiles in peripheral lymphocytes by arsenic exposure and skin lesion status in a Bangladeshi population. Cancer Epidemiol Biomarkers Prev. 2006;15:1367–1375. doi: 10.1158/1055-9965.EPI-06-0106. [DOI] [PubMed] [Google Scholar]
  8. Benbrahim-Tallaa L, Waalkes M. Inorganic arsenic and human prostate cancer. Environ Health Perspect. 2008;116:158–164. doi: 10.1289/ehp.10423. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Bodwell JE, Gosse JA, Nomikos AP, Hamilton JW. Arsenic disruption of steroid receptor gene activation: Complex dose-response effects are shared by several steroid receptors. Chem Res Toxicol. 2006;19:1619–1629. doi: 10.1021/tx060122q. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Bredfeldt TG, Kopplin MJ, Gandolfi AJ. Effects of arsenite on UROtsa cells: low-level arsenite causes accumulation of ubiquitinated proteins that is enhanced by reduction in cellular glutathione levels. Toxicology and Applied Pharmacology. 2004;198:412–418. doi: 10.1016/j.taap.2003.10.013. [DOI] [PubMed] [Google Scholar]
  11. Chang S, Jin B, Youn P, Park C, Park J-D, Ryu D-Y. Arsenic-induced toxicity and the protective role of ascorbic acid in mouse testis. Toxicology and applied pharmacology. 2007;218:196–203. doi: 10.1016/j.taap.2006.11.009. [DOI] [PubMed] [Google Scholar]
  12. Chang W-C, Hung J-J. Functional role of post-translational modifications of Sp1 in tumorigenesis. Journal of Biomedical Science. 2012;19 doi: 10.1186/1423-0127-19-94. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Chen CJ, Hsueh YM, Lai MS, Shyu MP, Chen SY, Wu MM, Kuo TL, Tai TY. Increased prevalence of hypertension and long-term arsenic exposure. Hypertension. 1995;25:53–60. [PubMed] [Google Scholar]
  14. Chen Y, Parvez F, Gamble M, Islam T, Ahmed A, Argos M, Graziano JH, Ahsan H. Arsenic exposure at low-to-moderate levels and skin lesions, arsenic metabolism, neurological functions, and biomarkers for respiratory and cardiovascular diseases: review of recent findings from the Health Effects of Arsenic Longitudinal Study (HEALS) in Bangladesh. Toxicol Appl Pharmacol. 2009;239:184–192. doi: 10.1016/j.taap.2009.01.010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Chen Y, Wu F, Liu M, Parvez F, Slavkovich V, Eunus M, Ahmed A, Argos M, Islam T, Rakibuz-Zaman M, et al. A prospective study of arsenic exposure, arsenic methylation capacity, and risk of cardiovascular disease in Bangladesh. Environmental health perspectives. 2013;121:832–838. doi: 10.1289/ehp.1205797. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Chervona Y, Hall MN, Arita A, Wu F, Sun H, Tseng HC, Ali E, Uddin MN, Liu XH, Zoroddu MA, et al. Associations between Arsenic Exposure and Global Posttranslational Histone Modifications among Adults in Bangladesh. Cancer Epidemiol Biomarkers Prev. 2012;21:2252–2260. doi: 10.1158/1055-9965.EPI-12-0833. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Davey JC, Bodwell JE, Gosse JA, Hamilton JW. Arsenic as an endocrine disruptor: Effects of arsenic on estrogen receptor-mediated gene expression in vivo and in cell culture. Toxicol Sci. 2007;98:75–86. doi: 10.1093/toxsci/kfm013. [DOI] [PubMed] [Google Scholar]
  18. De Coster S, van Leeuwen D, Jennen D, Koppen G, Den Hond E, Nelen V, Schoeters G, Baeyens W, van Delft J, Kleinjans J, et al. Gender-specific transcriptomic response to environmental exposure in Flemish adults. Environ Mol Mutagen. 2013;54:574–588. doi: 10.1002/em.21774. [DOI] [PubMed] [Google Scholar]
  19. Draghici S, Khatri P, Martins RP, Ostermeier GC, Krawetz SA. Global functional profiling of gene expression. Genomics. 2003;81:98–104. doi: 10.1016/s0888-7543(02)00021-6. [DOI] [PubMed] [Google Scholar]
  20. Flora SJS. Arsenic-induced oxidative stress and its reversibility. Free Radical Biology and Medicine. 2011;51:257–281. doi: 10.1016/j.freeradbiomed.2011.04.008. [DOI] [PubMed] [Google Scholar]
  21. Hartwig A, Groblinghoff UD, Beyersmann D, Natarajan AT, Filon R, Mullenders LH. Interaction of arsenic(III) with nucleotide excision repair in UV- irradiated human fibroblasts. Carcinogenesis. 1997;18:399–405. doi: 10.1093/carcin/18.2.399. [DOI] [PubMed] [Google Scholar]
  22. Hebels D, Jennen D, van Herwijnen M, Moonen E, Pedersen M, Knudsen L, Kleinjans J, de Kok TMC. Whole-genome gene expression modifications associated with nitrosamine exposure and micronucleus frequency in human blood cells. Mutagenesis. 2011;26:753–761. doi: 10.1093/mutage/ger043. [DOI] [PubMed] [Google Scholar]
  23. Hsieh F-I, Hwang T-S, Hsieh Y-C, Lo H-C, Su C-T, Hsu H-S, Chiou H-Y, Chen C-J. Risk of erectile dysfunction induced by arsenic exposure through well water consumption in Taiwan. Environmental health perspectives. 2008;116:532–536. doi: 10.1289/ehp.10930. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. IARC. Some drinking-water disinfectants and contaminants, including arsenic. Monographs on chloramine, chloral and chloral hydrate, dichloroacetic acid, trichloroacetic acid and 3-chloro-4-(dichloromethyl)-5-hydroxy-2(5H)-furanone. IARC Monogr Eval Carcinog Risks Hum. 2004;84:269–477. [PMC free article] [PubMed] [Google Scholar]
  25. Jana K, Jana S, Samanta P. Effects of chronic exposure to sodium arsenite on hypothalamo-pituitary-testicular activities in adult rats: possible an estrogenic mode of action. Reproductive biology and endocrinology. 2006;4:9. doi: 10.1186/1477-7827-4-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Kirkpatrick DS, Dale KV, Catania JM, Gandolfi AJ. Low-level arsenite causes accumulation of ubiquitinated proteins in rabbit renal cortical slices and HEK293 cells. Toxicology and Applied Pharmacology. 2003;186:101–109. doi: 10.1016/s0041-008x(02)00019-4. [DOI] [PubMed] [Google Scholar]
  27. Lampe JW, Stepaniants SB, Mao M, Radich JP, Dai H, Linsley PS, Friend SH, Potter JD. Signatures of environmental exposures using peripheral leukocyte gene expression: tobacco smoke. Cancer Epidemiol Biomarkers Prev. 2004;13:445–453. [PubMed] [Google Scholar]
  28. Leek JT, Storey JD. Capturing heterogeneity in gene expression studies by surrogate variable analysis. PLoS Genet. 2007;3:1724–1735. doi: 10.1371/journal.pgen.0030161. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Li L, He SH, Sun JM, Davie JR. Gene regulation by Sp1 and Sp3. Biochemistry and Cell Biology-Biochimie Et Biologie Cellulaire. 2004;82:460–471. doi: 10.1139/o04-045. [DOI] [PubMed] [Google Scholar]
  30. Lu T, Liu J, LeCluyse E, Zhou Y, Cheng M, Waalkes M. Application of cDNA microarray to the study of arsenic-induced liver diseases in the population of Guizhou, China. Toxicological sciences : an official journal of the Society of Toxicology. 2001;59:185–192. doi: 10.1093/toxsci/59.1.185. [DOI] [PubMed] [Google Scholar]
  31. Meinel S, Ruhs S, Schumann K, Straetz N, Trenkmann K, Schreier B, Grosse I, Keilwagen J, Gekle M, Grossmann C. Mineralocorticoid receptor interaction with SP1 generates a new response element for pathophysiologically relevant gene expression. Nucleic Acids Research. 2013;41:8045–8060. doi: 10.1093/nar/gkt581. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Moeller G, Adamski J. Multifunctionality of human 17beta-hydroxysteroid dehydrogenases. Molecular and cellular endocrinology. 2006;248:47–55. doi: 10.1016/j.mce.2005.11.031. [DOI] [PubMed] [Google Scholar]
  33. Nixon DE, Mussmann GV, Eckdahl SJ, Moyer TP. Total Arsenic In Urine -Palladium Persulfate Vs Nickel As A Matrix Modifier For Graphite-Furnace Atomic-Absorption Spectrophotometry. Clinical Chemistry. 1991;37:1575–1579. [PubMed] [Google Scholar]
  34. Ou X-M, Chen K, Shih JC. Glucocorticoid and androgen activation of monoamine oxidase A is regulated differently by R1 and Sp1. Journal of Biological Chemistry. 2006;281:21512–21525. doi: 10.1074/jbc.M600250200. [DOI] [PubMed] [Google Scholar]
  35. Pan W-C, Seow W, Kile M, Hoffman E, Quamruzzaman Q, Rahman M, Mahiuddin G, Mostofa G, Lu Q, Christiani D. Association of low to moderate levels of arsenic exposure with risk of type 2 diabetes in Bangladesh. American journal of epidemiology. 2013;178:1563–1570. doi: 10.1093/aje/kwt195. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Pant N, Murthy R, Srivastava S. Male reproductive toxicity of sodium arsenite in mice. Human & experimental toxicology. 2004;23:399–403. doi: 10.1191/0960327104ht467oa. [DOI] [PubMed] [Google Scholar]
  37. Piccolo SR, Sun Y, Campbell JD, Lenburg ME, Bild AH, Johnson WE. A single-sample microarray normalization method to facilitate personalized-medicine workflows. Genomics. 2012;100:337–344. doi: 10.1016/j.ygeno.2012.08.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Pilsner J, Hall M, Liu X, Ilievski V, Slavkovich V, Levy D, Factor-Litvak P, Yunus M, Rahman M, Graziano J, et al. Influence of prenatal arsenic exposure and newborn sex on global methylation of cord blood DNA. PloS one. 2012;7 doi: 10.1371/journal.pone.0037147. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Porter W, Saville B, Hoivik D, Safe S. Functional synergy between the transcription factor Sp1 and the estrogen receptor. Molecular Endocrinology. 1997;11:1569–1580. doi: 10.1210/mend.11.11.9916. [DOI] [PubMed] [Google Scholar]
  40. Rahman M, Ng J, Naidu R. Chronic exposure of arsenic via drinking water and its adverse health impacts on humans. Environ Geochem Health. 2009;31(Suppl 1):189–200. doi: 10.1007/s10653-008-9235-0. [DOI] [PubMed] [Google Scholar]
  41. Ravenscroft P, Brammer H, Richards K. Arsenic pollution: a global synthesis. U.K.: Wiley-Blackwell; 2009. [Google Scholar]
  42. Reichard JF, Puga A. Effects of arsenic exposure on DNA methylation and epigenetic gene regulation. Epigenomics. 2010;2:87–104. doi: 10.2217/epi.09.45. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Ren XF, McHale CM, Skibola CF, Smith AH, Smith MT, Zhang LP. An Emerging Role for Epigenetic Dysregulation in Arsenic Toxicity and Carcinogenesis. Environmental Health Perspectives. 2011;119:11–19. doi: 10.1289/ehp.1002114. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Rossman TG, Uddin AN, Burns FJ. Evidence that arsenite acts as a cocarcinogen in skin cancer. Toxicology and Applied Pharmacology. 2004;198:394–404. doi: 10.1016/j.taap.2003.10.016. [DOI] [PubMed] [Google Scholar]
  45. Schuhmacher-Wolz U, Dieter H, Klein D, Schneider K. Oral exposure to inorganic arsenic: evaluation of its carcinogenic and non-carcinogenic effects. Crit Rev Toxicol. 2009;39:271–298. doi: 10.1080/10408440802291505. [DOI] [PubMed] [Google Scholar]
  46. Smyth GK. Linear models and empirical Bayes methods for assessing differential expression in microarray experiments. Statistical Applications in Genetics and Molecular Biology. 2004;3 doi: 10.2202/1544-6115.1027. Article 3. [DOI] [PubMed] [Google Scholar]
  47. Tseng CH, Chong CK, Tseng CP, Hsueh YM, Chiou HY, Tseng CC, Chen CJ. Long-term arsenic exposure and ischemic heart disease in arseniasis-hyperendemic villages in Taiwan. Toxicol Lett. 2003;137:15–21. doi: 10.1016/s0378-4274(02)00377-6. [DOI] [PubMed] [Google Scholar]
  48. van Leeuwen DM, Gottschalk RWH, van Herwijnen MH, Moonen EJ, Kleinjans JCS, van Delft JHM. Differential gene expression in human peripheral blood mononuclear cells induced by cigarette smoke and its constituents. Toxicol Sci. 2005;86:200–210. doi: 10.1093/toxsci/kfi168. [DOI] [PubMed] [Google Scholar]
  49. Waalkes M, Liu J, Chen H, Xie Y, Achanzar W, Zhou Y-S, Cheng M-L, Diwan B. Estrogen signaling in livers of male mice with hepatocellular carcinoma induced by exposure to arsenic in utero. J Natl Cancer Inst. 2004;96:466–474. doi: 10.1093/jnci/djh070. [DOI] [PubMed] [Google Scholar]
  50. Waalkes M, Ward J, Liu J, Diwan B. Transplacental carcinogenicity of inorganic arsenic in the drinking water: induction of hepatic, ovarian, pulmonary, and adrenal tumors in mice. Toxicology and applied pharmacology. 2003;186:7–17. doi: 10.1016/s0041-008x(02)00022-4. [DOI] [PubMed] [Google Scholar]
  51. Wang Z, Neuburg D, Li C, Su L, Kim J, Chen J, Christiani D. Global gene expression profiling in whole-blood samples from individuals exposed to metal fumes. Environmental health perspectives. 2005;113:233–241. doi: 10.1289/txg.7273. [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. Wu M-M, Chiou H-Y, Ho IC, Chen C-J, Lee T-C. Gene expression of inflammatory molecules in circulating lymphocytes from arsenic-exposed human subjects. Environmental health perspectives. 2003;111:1429–1438. doi: 10.1289/ehp.6396. [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Xu W, Bao H, Liu F, Liu L, Zhu Y-G, She J, Dong S, Cai M, Li L, Li C, et al. Environmental exposure to arsenic may reduce human semen quality: associations derived from a Chinese cross-sectional study. Environmental health : a global access science source. 2012;11:46. doi: 10.1186/1476-069X-11-46. [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Yang S-Y, He X-Y, Schulz H. Multiple functions of type 10 17beta-hydroxysteroid dehydrogenase. Trends in endocrinology and metabolism: TEM. 2005;16:167–175. doi: 10.1016/j.tem.2005.03.006. [DOI] [PubMed] [Google Scholar]
  55. Yuan HQ, Gong AY, Young CYF. Involvement of transcription factor Sp1 in quercetin-mediated inhibitory effect on the androgen receptor in human prostate cancer cells. Carcinogenesis. 2005;26:793–801. doi: 10.1093/carcin/bgi021. [DOI] [PubMed] [Google Scholar]
  56. Zhou X, Sun X, Mobarak C, Gandolfi AJ, Burchiel SW, Hudson LG, Liu KJ. Differential binding of monomethylarsonous acid compared to arsenite and arsenic trioxide with zinc finger peptides and proteins. Chemical Research in Toxicology. 2014;27:690–698. doi: 10.1021/tx500022j. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

NIHMS670646-supplement.pdf (1.1MB, pdf)

RESOURCES